The present study was approved by the Ethics Committee of the Affiliated Hospital of North Sichuan Medical College (Nanchong, China; approval no. 2020ER(A)024; May 29, 2019). Written informed consent was provided by patients, their authorized agents or their close relatives prior to the study start for use of their tissue samples and clinical details.

A total of 139 patients with HCC from the Affiliated Hospital of North Sichuan Medical College formed the study cohort, between June 2014 and December 2016. The present study included 114 men and 25 women with HCC, with an average age of 54.22 years (age range, 36–73 years). HCC was confirmed via histological analysis. A total of 139 HCC tissues and 81 paired para-tumor tissues were selected from patients who had not undergone radiotherapy or chemotherapy prior to surgery. In addition, patients with distant metastasis, Child-Pugh (28) liver function of grade C or other types of cancer were excluded from the present study. The cancer tissues and corresponding adjacent tissues were applied to produce tissue microarrays (TMAs).

Survival analysis data was obtained via the telephone, the follow-up visit comprised liver-function tests, chest radiography and measurement of α-fetoprotein (AFP) expression. Follow-up was at least every 2 months in the first 6 months following surgery, every 3 months for 6 months-2 years after surgery, and every 6 months for 2–5 years following surgery. If necessary, computed tomography or magnetic resonance imaging were also performed to diagnose tumor recurrence. Postoperative adjuvant therapy and treatment of tumor recurrence were communicated by our multidisciplinary team. Surpassing 5 years post-surgery, the survival of patients with HCC was largely affected by several clinically unrelated factors, for example household income (3), thus the survival of all patients was calculated up to 5 years post-surgery.

Fresh tissue samples were fixed in 4% formalin for 24 h at room temperature, washed five times with PBS (3 min each) and subsequently dehydrated with ethanol. Tissue samples were embedded in paraffin and cut into 4-µm-thick sections (diameter of 2 mm) to produce TMAs. TMAs were deparaffinized in xylene and rehydrated in a descending ethanol series (100, 95 and 85%) at room temperature. TMAs were incubated with 0.3% H₂O₂ to inhibit endogenous peroxidase activity, and antigen-retrieval was subsequently performed with ethylenediamine tetraacetic acid antigen-retrieval solution (Beyotime Institute of Biotechnology; P0085), using a microwave for 15 min. TMAs were blocked with 10% goat serum (Beijing Solarbio Science & Technology Co., Ltd.; SL038) for 30 min at room temperature, and incubated with anti-FBL (1:100; Abcam; cat. no. ab166630) overnight at 4°C. Following the primary incubation, TMAs were incubated with a secondary antibody for 1 h at 37°C (1:1; Dako; Agilent Technologies, Inc; cat. no. K5007). The REAL™ EnVision™ immunohistochemistry kit include the secondary antibody, and the secondary antibody from the manufacturer has come out of the working fluid concentration. The TAMs were color rendered using the REAL™ EnVision™ immunohistochemistry kit (Dako; Agilent Technologies, Inc; cat. no. K5007), according to the manufacturer's instructions. To score the expression of FBL in the TMAs, the TMAs were imaged using an automatic immunohistochemical section scanning system (Pannorramic SCAN, Hungary).

Color intensity was divided into four classifications of staining, as follows: None, score=0; weak, score=1; moderate, score=2 and intense, score=3. The proportion of positive cells was ranked into five levels, as follows: <5%, score=0; 5–25%, score=1; 26–50%, score=2; 51–74%, score=3 and ≥75%, score=4. Multiplication of the staining score with the score for the percentage of positive cells provided the final score. The cut-off point for high FBL expression was based on the median IHC score −6 (29). Thus, a score <6 was classified as low FBL protein expression, while a score ≥6 was classified as high FBL protein expression. Each score was independently evaluated by two experienced pathologists from the Department of Pathology, The First People's Hospital of Neijiang (Neijiang, China). In the case of ambiguity, a third experienced pathologist from the Department of Pathology, The First People's Hospital of Neijiang (Neijiang, China) judged the final outcome.

The UALCAN database (ualcan.path.uab.edu) was used to analyze FBL expression in HCC, the methylation level of its promoters and identify the associated genes, which was downloaded from The Cancer Genome Atlas (TCGA) RNA-sequencing data (ualcan.path.uab.edu). The statistical tests were performed via TCGA analysis of UALCAN (Figs. 1A and B, and 3B). FBL expression was assessed in HCC from 50 normal liver tissues and 371 primary tumor tissues. Transcripts per million was used to measure expression. Functional enrichment analysis of the associated genes was performed using the Database for Annotation, Visualization and Integrated Discovery (DAVID 6.7, David.ncifcrf.gov). The Gene Ontology (GO) database (David.ncifcrf.gov) was used to assess 186 FBL-related genes (Person-CC≥0.5). The mutation or amplification of FBL was validated using cBioportal (www.cbioportal.org), which involved four datasets, as follows: INSERM, Nat Genet 2015, 243 samples (30); AMC, Hepatology 2014, 231 samples (31); TCGA, Firehorse Legacy, 442 samples (http://gdac.broadinstitute.org/runs/stddata), Pancancer Atlas, 372 samples (32–41). The micro (mi)RNAs that bind with the 3′-untranslated region (UTR) of FBL were predicted using TargetScan 3.1 (www.targetscan.org). The methylation of promoters was assessed within UALCAN using TCGA RNA-sequencing data. The β-value indicated the level of DNA methylation ranging from 0 (unmethylated) to 1 (fully methylated).

Statistical analysis were performed using SPSS v23 software (IBM Corp.). The χ² test was for categorical data, as well as to determine the association between FBL protein expression and the clinicopathological characteristics of patients with HCC. Survival analysis was performed using the Kaplan-Meier method and log-rank test. Univariate and multivariate analyses was performed using the Cox proportional hazards model to determine the prognostic values of the risk factors associated with HCC. P<0.05 was considered to indicate a statistically significant difference.

Analyses of the UALCAN database demonstrated that FBL mRNA expression was higher in HCC tissues compared with normal liver tissues; 50 normal tissues and 371 primary tumor tissues were assessed (Fig. 1A). In addition, high FBL mRNA expression was observed in patients with advanced HCC than those with early HCC (Fig. 1B). A total of 139 tumor tissues and 81 paired para-tumor tissues were used in the present study to assess FBL protein expression. IHC analysis was undertaken by our research team and the results demonstrated that FBL protein was predominantly located in the nuclei, at significantly higher levels in HCC tissues (Fig. 1C; upper panel, para-tumor tissues; lower panel, tumor tissues). Among the 139 tumor tissues, IHC analysis revealed 64 patients (46.0%) with high FBL expression. However, among the 81 para-tumor tissues, 11 patients (13.6%) had high FBL expression (P<0.001; Fig. 1D).

The study cohort comprised 114 men and 25 women with HCC, and the average age was 54.22 years (Table I). The χ² test was used to assess the association between FBL expression and the clinicopathological characteristics of patients with HCC. The results demonstrated that high FBL expression was significantly associated with larger tumor diameter (P<0.001) and advanced TNM stage (P=0.003). However, no significant associations were observed between FBL expression and age, sex, tumor number, tumor differentiation, infection with the hepatitis-B virus, liver cirrhosis, AFP level and Child-Pugh grade (Table I).

Kaplan-Meier survival analysis was performed to assess the effect of FBL protein expression on the survival of patients with HCC. The results demonstrated that overall survival time (P=0.032) and disease-free survival time (P=0.004) were significantly shorter in the high FBL expression group compared with the low FBL expression group (Fig. 2). Cox regression analysis was subsequently performed to determine the prognostic values of the risk factors in patients with HCC. As presented in Table II, univariate analysis demonstrated that high FBL expression (P=0.036), tumor diameter (P<0.001) and TNM stage (P<0.001) predicted a shorter overall survival time, while high FBL expression (P=0.006), tumor diameter (P0.001), TNM stage (P<0.001) and tumor number (P=0.019) predicted a shorter disease-free survival time. Notably, multivariate analysis demonstrated that high FBL expression was not an independent risk factor for overall survival and disease-free survival time, suggesting that FBL mainly affects the prognosis of patients with HCC by promoting tumor progression (Table II).

The UALCAN database was used to identify genes associated with FBL. The results demonstrated that 186 genes (Person-CC≥0.5) were significantly associated with the FBL gene. Functional enrichment analysis using the DAVID and GO databases demonstrated a majority of biological processes and signaling pathways associated with cancer progression, such as ‘cell-cell adhesion’, ‘G1/S transition of mitotic cell cycle’ and ‘positive regulation of epidermal growth factor receptor signaling pathway’. The notable cancer-associated biological processes are presented in Fig. 3A. The molecular mechanisms regulating FBL expression are poorly understood (21,42). Using publicly available data, it was demonstrated that methylation of the FBL promoter was high (Fig. 3B); however, the frequency of amplification and mutation was notably low in HCC (Fig. 3C). The miRNAs that may bind with the 3′-UTR of the FBL gene were predicted using TargetScan; however, no conserved miRNAs were identified for FBL. Taken together, these results suggest that FBL expression is mainly regulated by methylation.